
\begin{table}[htbp]
   \caption{\label{tab:mid_persistence} Fatal MID Initiation: Persistent Instability?}
   \centering
   \begin{tabular}{lcccc}
      \tabularnewline \midrule \midrule
      Dependent Variable: & \multicolumn{4}{c}{Fatal MID onset $\times 100$}\\
      Model:                                    & (1)            & (2)            & (3)            & (4)\\  
      \midrule
      \emph{Variables}\\
      Lost home rule \& lost unity              & 0.1324$^{***}$ & 0.1944$^{***}$ & 0.1887$^{***}$ &   \\   
                                                & (0.0255)       & (0.0341)       & (0.0323)       &   \\   
      Lost home rule \& lost unity (post-1816)  &                &                &                & 0.1733$^{***}$\\   
                                                &                &                &                & (0.0262)\\   
      Lost home rule \& lost unity (pre-1816)   &                &                &                & 0.2060$^{**}$\\   
                                                &                &                &                & (0.0736)\\   
      Lost unity only                           & 0.0453         & 0.0570         & 0.0530         &   \\   
                                                & (0.0456)       & (0.0443)       & (0.0418)       &   \\   
      Lost unity only (post-1816)               &                &                &                & 0.0209\\   
                                                &                &                &                & (0.0471)\\   
      Lost unity only (pre-1816)                &                &                &                & 0.0975$^{*}$\\   
                                                &                &                &                & (0.0383)\\   
      No. of Wars  (1400-1790, log)             & 0.0191         & 0.0188         & -0.0095        & -0.0098\\   
                                                & (0.0254)       & (0.0207)       & (0.0153)       & (0.0155)\\   
      \midrule
      \emph{Fixed-effects}\\
      Baseline controls                         & Yes            & Yes            & Yes            & Yes\\  
      Extended controls                         &                &                & Yes            & Yes\\  
      Year                                      & Yes            & Yes            & Yes            & Yes\\  
      Peace year                                & Yes            & Yes            & Yes            & Yes\\  
      State A                                   &                & Yes            & Yes            & Yes\\  
      State B                                   &                & Yes            & Yes            & Yes\\  
      Border duration A                         &                &                & Yes            & Yes\\  
      Border duration B                         &                &                & Yes            & Yes\\  
      \midrule
      \emph{Fit statistics}\\
      Observations                              & 188,210        & 188,210        & 188,210        & 188,210\\  
      \midrule \midrule
      \multicolumn{5}{l}{\emph{Clustered (Directed Dyad ID \& State A \& State B) standard-errors in parentheses}}\\
      \multicolumn{5}{l}{\emph{Signif. Codes: ***: 0.001, **: 0.01, *: 0.05, +: 0.1}}\\
   \end{tabular}
   
   \par \raggedright 
   \parbox[t]{\width=\textwidth}{\textbf{Notes:} OLS estimates of fatal MID iniation. 
          The unit of analysis is the directed country dyad year. 
          Baseline controls: relative territorial size of state A vs. state B; logged absolute size of country B; 
          Share of aggregate group governing in state A located in state B;
          Share of aggregate group governing in state A located in own country;
          Dummies for peace and calendar years.  
          Extended controls: logged aggregate group size of governing segments in A and B;
          ethnic fractionalization of countries A and B;
          conflict history (number of past years with ongoing MIDs involving A and B);
          time since last border change involving country A or B (FE). 
          Standard errors clustered on dyad, state A, and state B in parentheses.
          Significance codes: ***: 0.001, **: 0.01, *: 0.05, +: 0.1}
\end{table}


